The past decades have seen a proliferation of techniques for studying the functioning of the living human brain, and one area of neuroscience that has particularly benefited is the study of language. Methods such as PET, EEG, MEG, single cell recording and cortical stimulation mapping (CSM) are providing increasing insight about language function. However, each method shows a different aspect of functional activation, often leading to seemingly contradictory measurements. What is needed are methods for integrating and visualizing these diverse forms of data as a first step towards computational models of language function. Yet it is impossible and redundant for a single lab to collect all the data themselves in order to make these comparisons, and it is currently very difficult to obtain comparable data from other labs, both because of technical factors such as anatomical variation and non-standardization of language task protocols, and because of sociological factors relating to privacy and intellectual property.In this application we address these needs by extending and generalizing methods we have developed to organize neurosurgical language data for a single lab around a neuroanatomical framework. We will 1) develop a spatio-temporal experiment management system (STEMS) that handles complex multi-resolution brain map data, 2) develop methods for both humans and computers to interact with the database, including 3-D visualization of spatio-temporal results, and 3) develop methods for peer-to-peer sharing of independently maintained spatio-temporal experiment management systems.Our hypothesis is that this "bottom-up" approach of developing STEMSs for individual labs, then interconnecting them in a peer-to-peer database, is a viable approach to achieving a goal of the Human Brain Project to develop interoperable databases of raw neuroscience data. We will test this hypothesis by building at least two such STEMSs, and assessing their utility, both alone and in combination, in generating neuroscience results that would not have been easily generated without them.